The current implementation of rtc-efi is expecting all the 4
time services GET{SET}_TIME{WAKEUP} must be supported by UEFI
firmware. As per the EFI_RT_PROPERTIES_TABLE, the platform
specific implementations can choose to enable selective time
services based on the RTC device capabilities.
This patch does the following changes to provide GET/SET RTC
services on platforms that do not support the WAKEUP feature.
1) Relax time services cap check when creating a platform device.
2) Clear RTC_FEATURE_ALARM bit in the absence of WAKEUP services.
3) Conditional alarm entries in '/proc/driver/rtc'.
Signed-off-by: Shanker Donthineni <[email protected]>
---
drivers/rtc/rtc-efi.c | 48 ++++++++++++++++++++++++-------------------
include/linux/efi.h | 3 ++-
2 files changed, 29 insertions(+), 22 deletions(-)
diff --git a/drivers/rtc/rtc-efi.c b/drivers/rtc/rtc-efi.c
index e991cccdb6e9c..1e8bc6cc1e12d 100644
--- a/drivers/rtc/rtc-efi.c
+++ b/drivers/rtc/rtc-efi.c
@@ -188,9 +188,10 @@ static int efi_set_time(struct device *dev, struct rtc_time *tm)
static int efi_procfs(struct device *dev, struct seq_file *seq)
{
- efi_time_t eft, alm;
- efi_time_cap_t cap;
- efi_bool_t enabled, pending;
+ efi_time_t eft, alm;
+ efi_time_cap_t cap;
+ efi_bool_t enabled, pending;
+ struct rtc_device *rtc = dev_get_drvdata(dev);
memset(&eft, 0, sizeof(eft));
memset(&alm, 0, sizeof(alm));
@@ -213,23 +214,25 @@ static int efi_procfs(struct device *dev, struct seq_file *seq)
/* XXX fixme: convert to string? */
seq_printf(seq, "Timezone\t: %u\n", eft.timezone);
- seq_printf(seq,
- "Alarm Time\t: %u:%u:%u.%09u\n"
- "Alarm Date\t: %u-%u-%u\n"
- "Alarm Daylight\t: %u\n"
- "Enabled\t\t: %s\n"
- "Pending\t\t: %s\n",
- alm.hour, alm.minute, alm.second, alm.nanosecond,
- alm.year, alm.month, alm.day,
- alm.daylight,
- enabled == 1 ? "yes" : "no",
- pending == 1 ? "yes" : "no");
-
- if (eft.timezone == EFI_UNSPECIFIED_TIMEZONE)
- seq_puts(seq, "Timezone\t: unspecified\n");
- else
- /* XXX fixme: convert to string? */
- seq_printf(seq, "Timezone\t: %u\n", alm.timezone);
+ if (test_bit(RTC_FEATURE_ALARM, rtc->features)) {
+ seq_printf(seq,
+ "Alarm Time\t: %u:%u:%u.%09u\n"
+ "Alarm Date\t: %u-%u-%u\n"
+ "Alarm Daylight\t: %u\n"
+ "Enabled\t\t: %s\n"
+ "Pending\t\t: %s\n",
+ alm.hour, alm.minute, alm.second, alm.nanosecond,
+ alm.year, alm.month, alm.day,
+ alm.daylight,
+ enabled == 1 ? "yes" : "no",
+ pending == 1 ? "yes" : "no");
+
+ if (eft.timezone == EFI_UNSPECIFIED_TIMEZONE)
+ seq_puts(seq, "Timezone\t: unspecified\n");
+ else
+ /* XXX fixme: convert to string? */
+ seq_printf(seq, "Timezone\t: %u\n", alm.timezone);
+ }
/*
* now prints the capabilities
@@ -269,7 +272,10 @@ static int __init efi_rtc_probe(struct platform_device *dev)
rtc->ops = &efi_rtc_ops;
clear_bit(RTC_FEATURE_UPDATE_INTERRUPT, rtc->features);
- set_bit(RTC_FEATURE_ALARM_WAKEUP_ONLY, rtc->features);
+ if (efi_rt_services_supported(EFI_RT_SUPPORTED_WAKEUP_SERVICES))
+ set_bit(RTC_FEATURE_ALARM_WAKEUP_ONLY, rtc->features);
+ else
+ clear_bit(RTC_FEATURE_ALARM, rtc->features);
device_init_wakeup(&dev->dev, true);
diff --git a/include/linux/efi.h b/include/linux/efi.h
index 4b27519143f56..98598bd1d2fa5 100644
--- a/include/linux/efi.h
+++ b/include/linux/efi.h
@@ -668,7 +668,8 @@ extern struct efi {
#define EFI_RT_SUPPORTED_ALL 0x3fff
-#define EFI_RT_SUPPORTED_TIME_SERVICES 0x000f
+#define EFI_RT_SUPPORTED_TIME_SERVICES 0x0003
+#define EFI_RT_SUPPORTED_WAKEUP_SERVICES 0x000c
#define EFI_RT_SUPPORTED_VARIABLE_SERVICES 0x0070
extern struct mm_struct efi_mm;
--
2.25.1
On Tue, 27 Dec 2022 at 05:09, Shanker Donthineni <[email protected]> wrote:
>
> The current implementation of rtc-efi is expecting all the 4
> time services GET{SET}_TIME{WAKEUP} must be supported by UEFI
> firmware. As per the EFI_RT_PROPERTIES_TABLE, the platform
> specific implementations can choose to enable selective time
> services based on the RTC device capabilities.
>
> This patch does the following changes to provide GET/SET RTC
> services on platforms that do not support the WAKEUP feature.
>
> 1) Relax time services cap check when creating a platform device.
> 2) Clear RTC_FEATURE_ALARM bit in the absence of WAKEUP services.
> 3) Conditional alarm entries in '/proc/driver/rtc'.
>
> Signed-off-by: Shanker Donthineni <[email protected]>
Queued as a fix in efi/urgent, thanks.
> ---
> drivers/rtc/rtc-efi.c | 48 ++++++++++++++++++++++++-------------------
> include/linux/efi.h | 3 ++-
> 2 files changed, 29 insertions(+), 22 deletions(-)
>
> diff --git a/drivers/rtc/rtc-efi.c b/drivers/rtc/rtc-efi.c
> index e991cccdb6e9c..1e8bc6cc1e12d 100644
> --- a/drivers/rtc/rtc-efi.c
> +++ b/drivers/rtc/rtc-efi.c
> @@ -188,9 +188,10 @@ static int efi_set_time(struct device *dev, struct rtc_time *tm)
>
> static int efi_procfs(struct device *dev, struct seq_file *seq)
> {
> - efi_time_t eft, alm;
> - efi_time_cap_t cap;
> - efi_bool_t enabled, pending;
> + efi_time_t eft, alm;
> + efi_time_cap_t cap;
> + efi_bool_t enabled, pending;
> + struct rtc_device *rtc = dev_get_drvdata(dev);
>
> memset(&eft, 0, sizeof(eft));
> memset(&alm, 0, sizeof(alm));
> @@ -213,23 +214,25 @@ static int efi_procfs(struct device *dev, struct seq_file *seq)
> /* XXX fixme: convert to string? */
> seq_printf(seq, "Timezone\t: %u\n", eft.timezone);
>
> - seq_printf(seq,
> - "Alarm Time\t: %u:%u:%u.%09u\n"
> - "Alarm Date\t: %u-%u-%u\n"
> - "Alarm Daylight\t: %u\n"
> - "Enabled\t\t: %s\n"
> - "Pending\t\t: %s\n",
> - alm.hour, alm.minute, alm.second, alm.nanosecond,
> - alm.year, alm.month, alm.day,
> - alm.daylight,
> - enabled == 1 ? "yes" : "no",
> - pending == 1 ? "yes" : "no");
> -
> - if (eft.timezone == EFI_UNSPECIFIED_TIMEZONE)
> - seq_puts(seq, "Timezone\t: unspecified\n");
> - else
> - /* XXX fixme: convert to string? */
> - seq_printf(seq, "Timezone\t: %u\n", alm.timezone);
> + if (test_bit(RTC_FEATURE_ALARM, rtc->features)) {
> + seq_printf(seq,
> + "Alarm Time\t: %u:%u:%u.%09u\n"
> + "Alarm Date\t: %u-%u-%u\n"
> + "Alarm Daylight\t: %u\n"
> + "Enabled\t\t: %s\n"
> + "Pending\t\t: %s\n",
> + alm.hour, alm.minute, alm.second, alm.nanosecond,
> + alm.year, alm.month, alm.day,
> + alm.daylight,
> + enabled == 1 ? "yes" : "no",
> + pending == 1 ? "yes" : "no");
> +
> + if (eft.timezone == EFI_UNSPECIFIED_TIMEZONE)
> + seq_puts(seq, "Timezone\t: unspecified\n");
> + else
> + /* XXX fixme: convert to string? */
> + seq_printf(seq, "Timezone\t: %u\n", alm.timezone);
> + }
>
> /*
> * now prints the capabilities
> @@ -269,7 +272,10 @@ static int __init efi_rtc_probe(struct platform_device *dev)
>
> rtc->ops = &efi_rtc_ops;
> clear_bit(RTC_FEATURE_UPDATE_INTERRUPT, rtc->features);
> - set_bit(RTC_FEATURE_ALARM_WAKEUP_ONLY, rtc->features);
> + if (efi_rt_services_supported(EFI_RT_SUPPORTED_WAKEUP_SERVICES))
> + set_bit(RTC_FEATURE_ALARM_WAKEUP_ONLY, rtc->features);
> + else
> + clear_bit(RTC_FEATURE_ALARM, rtc->features);
>
> device_init_wakeup(&dev->dev, true);
>
> diff --git a/include/linux/efi.h b/include/linux/efi.h
> index 4b27519143f56..98598bd1d2fa5 100644
> --- a/include/linux/efi.h
> +++ b/include/linux/efi.h
> @@ -668,7 +668,8 @@ extern struct efi {
>
> #define EFI_RT_SUPPORTED_ALL 0x3fff
>
> -#define EFI_RT_SUPPORTED_TIME_SERVICES 0x000f
> +#define EFI_RT_SUPPORTED_TIME_SERVICES 0x0003
> +#define EFI_RT_SUPPORTED_WAKEUP_SERVICES 0x000c
> #define EFI_RT_SUPPORTED_VARIABLE_SERVICES 0x0070
>
> extern struct mm_struct efi_mm;
> --
> 2.25.1
>
Hi Alexandre Belloni,
On 1/2/23 16:21, Alexandre Belloni wrote:
> On 02/01/2023 11:47:11+0100, Ard Biesheuvel wrote:
>> On Tue, 27 Dec 2022 at 05:09, Shanker Donthineni<[email protected]> wrote:
>>> The current implementation of rtc-efi is expecting all the 4
>>> time services GET{SET}_TIME{WAKEUP} must be supported by UEFI
>>> firmware. As per the EFI_RT_PROPERTIES_TABLE, the platform
>>> specific implementations can choose to enable selective time
>>> services based on the RTC device capabilities.
>>>
>>> This patch does the following changes to provide GET/SET RTC
>>> services on platforms that do not support the WAKEUP feature.
>>>
>>> 1) Relax time services cap check when creating a platform device.
>>> 2) Clear RTC_FEATURE_ALARM bit in the absence of WAKEUP services.
>>> 3) Conditional alarm entries in '/proc/driver/rtc'.
>>>
>>> Signed-off-by: Shanker Donthineni<[email protected]>
>> Queued as a fix in efi/urgent, thanks.
> This rather seems like an rtc heavy patch and the subject line is
> misleading. This should be rtc: efi:
> Also, I'm pretty sure this doesn't qualify as an urgent fix.
>
Thanks, I'll post v2 patch with your suggested subject.
On 02/01/2023 11:47:11+0100, Ard Biesheuvel wrote:
> On Tue, 27 Dec 2022 at 05:09, Shanker Donthineni <[email protected]> wrote:
> >
> > The current implementation of rtc-efi is expecting all the 4
> > time services GET{SET}_TIME{WAKEUP} must be supported by UEFI
> > firmware. As per the EFI_RT_PROPERTIES_TABLE, the platform
> > specific implementations can choose to enable selective time
> > services based on the RTC device capabilities.
> >
> > This patch does the following changes to provide GET/SET RTC
> > services on platforms that do not support the WAKEUP feature.
> >
> > 1) Relax time services cap check when creating a platform device.
> > 2) Clear RTC_FEATURE_ALARM bit in the absence of WAKEUP services.
> > 3) Conditional alarm entries in '/proc/driver/rtc'.
> >
> > Signed-off-by: Shanker Donthineni <[email protected]>
>
> Queued as a fix in efi/urgent, thanks.
This rather seems like an rtc heavy patch and the subject line is
misleading. This should be rtc: efi:
Also, I'm pretty sure this doesn't qualify as an urgent fix.
>
>
> > ---
> > drivers/rtc/rtc-efi.c | 48 ++++++++++++++++++++++++-------------------
> > include/linux/efi.h | 3 ++-
> > 2 files changed, 29 insertions(+), 22 deletions(-)
> >
> > diff --git a/drivers/rtc/rtc-efi.c b/drivers/rtc/rtc-efi.c
> > index e991cccdb6e9c..1e8bc6cc1e12d 100644
> > --- a/drivers/rtc/rtc-efi.c
> > +++ b/drivers/rtc/rtc-efi.c
> > @@ -188,9 +188,10 @@ static int efi_set_time(struct device *dev, struct rtc_time *tm)
> >
> > static int efi_procfs(struct device *dev, struct seq_file *seq)
> > {
> > - efi_time_t eft, alm;
> > - efi_time_cap_t cap;
> > - efi_bool_t enabled, pending;
> > + efi_time_t eft, alm;
> > + efi_time_cap_t cap;
> > + efi_bool_t enabled, pending;
> > + struct rtc_device *rtc = dev_get_drvdata(dev);
> >
> > memset(&eft, 0, sizeof(eft));
> > memset(&alm, 0, sizeof(alm));
> > @@ -213,23 +214,25 @@ static int efi_procfs(struct device *dev, struct seq_file *seq)
> > /* XXX fixme: convert to string? */
> > seq_printf(seq, "Timezone\t: %u\n", eft.timezone);
> >
> > - seq_printf(seq,
> > - "Alarm Time\t: %u:%u:%u.%09u\n"
> > - "Alarm Date\t: %u-%u-%u\n"
> > - "Alarm Daylight\t: %u\n"
> > - "Enabled\t\t: %s\n"
> > - "Pending\t\t: %s\n",
> > - alm.hour, alm.minute, alm.second, alm.nanosecond,
> > - alm.year, alm.month, alm.day,
> > - alm.daylight,
> > - enabled == 1 ? "yes" : "no",
> > - pending == 1 ? "yes" : "no");
> > -
> > - if (eft.timezone == EFI_UNSPECIFIED_TIMEZONE)
> > - seq_puts(seq, "Timezone\t: unspecified\n");
> > - else
> > - /* XXX fixme: convert to string? */
> > - seq_printf(seq, "Timezone\t: %u\n", alm.timezone);
> > + if (test_bit(RTC_FEATURE_ALARM, rtc->features)) {
> > + seq_printf(seq,
> > + "Alarm Time\t: %u:%u:%u.%09u\n"
> > + "Alarm Date\t: %u-%u-%u\n"
> > + "Alarm Daylight\t: %u\n"
> > + "Enabled\t\t: %s\n"
> > + "Pending\t\t: %s\n",
> > + alm.hour, alm.minute, alm.second, alm.nanosecond,
> > + alm.year, alm.month, alm.day,
> > + alm.daylight,
> > + enabled == 1 ? "yes" : "no",
> > + pending == 1 ? "yes" : "no");
> > +
> > + if (eft.timezone == EFI_UNSPECIFIED_TIMEZONE)
> > + seq_puts(seq, "Timezone\t: unspecified\n");
> > + else
> > + /* XXX fixme: convert to string? */
> > + seq_printf(seq, "Timezone\t: %u\n", alm.timezone);
> > + }
> >
> > /*
> > * now prints the capabilities
> > @@ -269,7 +272,10 @@ static int __init efi_rtc_probe(struct platform_device *dev)
> >
> > rtc->ops = &efi_rtc_ops;
> > clear_bit(RTC_FEATURE_UPDATE_INTERRUPT, rtc->features);
> > - set_bit(RTC_FEATURE_ALARM_WAKEUP_ONLY, rtc->features);
> > + if (efi_rt_services_supported(EFI_RT_SUPPORTED_WAKEUP_SERVICES))
> > + set_bit(RTC_FEATURE_ALARM_WAKEUP_ONLY, rtc->features);
> > + else
> > + clear_bit(RTC_FEATURE_ALARM, rtc->features);
> >
> > device_init_wakeup(&dev->dev, true);
> >
> > diff --git a/include/linux/efi.h b/include/linux/efi.h
> > index 4b27519143f56..98598bd1d2fa5 100644
> > --- a/include/linux/efi.h
> > +++ b/include/linux/efi.h
> > @@ -668,7 +668,8 @@ extern struct efi {
> >
> > #define EFI_RT_SUPPORTED_ALL 0x3fff
> >
> > -#define EFI_RT_SUPPORTED_TIME_SERVICES 0x000f
> > +#define EFI_RT_SUPPORTED_TIME_SERVICES 0x0003
> > +#define EFI_RT_SUPPORTED_WAKEUP_SERVICES 0x000c
> > #define EFI_RT_SUPPORTED_VARIABLE_SERVICES 0x0070
> >
> > extern struct mm_struct efi_mm;
> > --
> > 2.25.1
> >
--
Alexandre Belloni, co-owner and COO, Bootlin
Embedded Linux and Kernel engineering
https://bootlin.com
On Mon, 2 Jan 2023 at 23:21, Alexandre Belloni
<[email protected]> wrote:
>
> On 02/01/2023 11:47:11+0100, Ard Biesheuvel wrote:
> > On Tue, 27 Dec 2022 at 05:09, Shanker Donthineni <[email protected]> wrote:
> > >
> > > The current implementation of rtc-efi is expecting all the 4
> > > time services GET{SET}_TIME{WAKEUP} must be supported by UEFI
> > > firmware. As per the EFI_RT_PROPERTIES_TABLE, the platform
> > > specific implementations can choose to enable selective time
> > > services based on the RTC device capabilities.
> > >
> > > This patch does the following changes to provide GET/SET RTC
> > > services on platforms that do not support the WAKEUP feature.
> > >
> > > 1) Relax time services cap check when creating a platform device.
> > > 2) Clear RTC_FEATURE_ALARM bit in the absence of WAKEUP services.
> > > 3) Conditional alarm entries in '/proc/driver/rtc'.
> > >
> > > Signed-off-by: Shanker Donthineni <[email protected]>
> >
> > Queued as a fix in efi/urgent, thanks.
>
> This rather seems like an rtc heavy patch and the subject line is
> misleading. This should be rtc: efi:
> Also, I'm pretty sure this doesn't qualify as an urgent fix.
>
I'm happy to drop it from my tree, but please add a cc:stable so it
gets backported to v6.1 at least. Otherwise, EFI compliant systems
that implement get/set_time but not get/set_wakeup_time have no RTC at
all on any LTS kernel until a year from now, and this was never the
intent when we introduced the EFI_RT_PROPERTIES_TABLE.
On Tue, 3 Jan 2023 at 17:37, Shanker Donthineni <[email protected]> wrote:
>
> Hi Ard Biesheuvel,
>
> On 1/3/23 03:18, Ard Biesheuvel wrote:
> > External email: Use caution opening links or attachments
> >
> >
> > On Mon, 2 Jan 2023 at 23:21, Alexandre Belloni
> > <[email protected]> wrote:
> >>
> >> On 02/01/2023 11:47:11+0100, Ard Biesheuvel wrote:
> >>> On Tue, 27 Dec 2022 at 05:09, Shanker Donthineni <[email protected]> wrote:
> >>>>
> >>>> The current implementation of rtc-efi is expecting all the 4
> >>>> time services GET{SET}_TIME{WAKEUP} must be supported by UEFI
> >>>> firmware. As per the EFI_RT_PROPERTIES_TABLE, the platform
> >>>> specific implementations can choose to enable selective time
> >>>> services based on the RTC device capabilities.
> >>>>
> >>>> This patch does the following changes to provide GET/SET RTC
> >>>> services on platforms that do not support the WAKEUP feature.
> >>>>
> >>>> 1) Relax time services cap check when creating a platform device.
> >>>> 2) Clear RTC_FEATURE_ALARM bit in the absence of WAKEUP services.
> >>>> 3) Conditional alarm entries in '/proc/driver/rtc'.
> >>>>
> >>>> Signed-off-by: Shanker Donthineni <[email protected]>
> >>>
> >>> Queued as a fix in efi/urgent, thanks.
> >>
> >> This rather seems like an rtc heavy patch and the subject line is
> >> misleading. This should be rtc: efi:
> >> Also, I'm pretty sure this doesn't qualify as an urgent fix.
> >>
> >
> > I'm happy to drop it from my tree, but please add a cc:stable so it
> > gets backported to v6.1 at least. Otherwise, EFI compliant systems
> > that implement get/set_time but not get/set_wakeup_time have no RTC at
> > all on any LTS kernel until a year from now, and this was never the
> > intent when we introduced the EFI_RT_PROPERTIES_TABLE.
>
> Thanks for considering the fix for stable releases, I'll post v3 patch
> with tag 'CC: <[email protected]> # v6.0+'
No please don't resend the patch
Hi Ard Biesheuvel,
On 1/3/23 03:18, Ard Biesheuvel wrote:
> External email: Use caution opening links or attachments
>
>
> On Mon, 2 Jan 2023 at 23:21, Alexandre Belloni
> <[email protected]> wrote:
>>
>> On 02/01/2023 11:47:11+0100, Ard Biesheuvel wrote:
>>> On Tue, 27 Dec 2022 at 05:09, Shanker Donthineni <[email protected]> wrote:
>>>>
>>>> The current implementation of rtc-efi is expecting all the 4
>>>> time services GET{SET}_TIME{WAKEUP} must be supported by UEFI
>>>> firmware. As per the EFI_RT_PROPERTIES_TABLE, the platform
>>>> specific implementations can choose to enable selective time
>>>> services based on the RTC device capabilities.
>>>>
>>>> This patch does the following changes to provide GET/SET RTC
>>>> services on platforms that do not support the WAKEUP feature.
>>>>
>>>> 1) Relax time services cap check when creating a platform device.
>>>> 2) Clear RTC_FEATURE_ALARM bit in the absence of WAKEUP services.
>>>> 3) Conditional alarm entries in '/proc/driver/rtc'.
>>>>
>>>> Signed-off-by: Shanker Donthineni <[email protected]>
>>>
>>> Queued as a fix in efi/urgent, thanks.
>>
>> This rather seems like an rtc heavy patch and the subject line is
>> misleading. This should be rtc: efi:
>> Also, I'm pretty sure this doesn't qualify as an urgent fix.
>>
>
> I'm happy to drop it from my tree, but please add a cc:stable so it
> gets backported to v6.1 at least. Otherwise, EFI compliant systems
> that implement get/set_time but not get/set_wakeup_time have no RTC at
> all on any LTS kernel until a year from now, and this was never the
> intent when we introduced the EFI_RT_PROPERTIES_TABLE.
Thanks for considering the fix for stable releases, I'll post v3 patch
with tag 'CC: <[email protected]> # v6.0+'